C#汉字转拼音功能类(页 1) - 编程语言 - 我爱电脑技术论坛打造最好的电脑自学交流论坛

小迷糊 发表于 2008-6-11 02:09

C#汉字转拼音功能类

此代码属改造而来，没有太深的技术难度，仅仅是排除了原来代码的错误、增加了一些常规需求。目前此类可以实现的功能如下：DN?za5e7e
1、对中文字符串进行拼音转换的基本功能
2、支持并可自定义拼音单词间的分隔符`9g7t{Y@O
3、支持并可自定义是否将拼音单词首字母大写
4、支持并可自定义多音词预定音U+Cp5K&BU
5、支持并可自定义转换过程中要排除解析的字符，例如中文标点 Oqh0D5]K7?@,D
输出示例：
$v$gU:N"g-V
引发 System.Web.UI.Control.Init 事件以对页进行初始化
Yin-Fa System.Web.UI.Control.Init Shi-Jian-Yi-Dui-Ye-Jin-Xing-Chu-Shi-Hua
h QWF B
Hello!重庆和深圳的特色各是什么呢？
Hello!Chong Qing He Shen Zhen De Te Se Ge Shi ShenMe Ne ？3j:NzPX8o

代码如下：+s/`LJh.L(q2|
/**//// <summary>*`;|;@1p0Z.W"GC
/// 汉字转拼音工具。
/// </summary>
public sealed class CHS2PinYin&k F&h Zq7\'o
...{
/**//// <summary>-N9Zs0XB@
/// 包含字符 ASC 码的整形数组。
/// </summary>z+Eh'M$_0^c
private static int[] pv = new int[] ...{ -20319, -20317, -20304, -20295, -20292, -20283, -20265, -20257, -20242, -20230, -20051, -20036, -20032, -20026, -20002, -19990, -19986, -19982, -19976, -19805, -19784, -19775, -19774, -19763, -19756, -19751, -19746, -19741, -19739, -19728, -19725, -19715, -19540, -19531, -19525, -19515, -19500, -19484, -19479, -19467, -19289, -19288, -19281, -19275, -19270, -19263, -19261, -19249, -19243, -19242, -19238, -19235, -19227, -19224, -19218, -19212, -19038, -19023, -19018, -19006, -19003, -18996, -18977, -18961, -18952, -18783, -18774, -18773, -18763, -18756, -18741, -18735, -18731, -18722, -18710, -18697, -18696, -18526, -18518, -18501, -18490, -18478, -18463, -18448, -18447, -18446, -18239, -18237, -18231, -18220, -18211, -18201, -18184, -18183, -18181, -18012, -17997, -17988, -17970, -17964, -17961, -17950, -17947, -17931, -17928, -17922, -17759, -17752, -17733, -17730, -17721, -17703, -17701, -17697, -17692, -17683, -17676, -17496, -17487, -17482, -17468, -17454, -17433, -17427, -17417, -17202, -17185, -16983, -16970, -16942, -16915, -16733, -16708, -16706, -16689, -16664, -16657, -16647, -16474, -16470, -16465, -16459, -16452, -16448, -16433, -16429, -16427, -16423, -16419, -16412, -16407, -16403, -16401, -16393, -16220, -16216, -16212, -16205, -16202, -16187, -16180, -16171, -16169, -16158, -16155, -15959, -15958, -15944, -15933, -15920, -15915, -15903, -15889, -15878, -15707, -15701, -15681, -15667, -15661, -15659, -15652, -15640, -15631, -15625, -15454, -15448, -15436, -15435, -15419, -15416, -15408, -15394, -15385, -15377, -15375, -15369, -15363, -15362, -15183, -15180, -15165, -15158, -15153, -15150, -15149, -15144, -15143, -15141, -15140, -15139, -15128, -15121, -15119, -15117, -15110, -15109, -14941, -14937, -14933, -14930, -14929, -14928, -14926, -14922, -14921, -14914, -14908, -14902, -14894, -14889, -14882, -14873, -14871, -14857, -14678, -14674, -14670, -14668, -14663, -14654, -14645, -14630, -14594, -14429, -14407, -14399, -14384, -14379, -14368, -14355, -14353, -14345, -14170, -14159, -14151, -14149, -14145, -14140, -14137, -14135, -14125, -14123, -14122, -14112, -14109, -14099, -14097, -14094, -14092, -14090, -14087, -14083, -13917, -13914, -13910, -13907, -13906, -13905, -13896, -13894, -13878, -13870, -13859, -13847, -13831, -13658, -13611, -13601, -13406, -13404, -13400, -13398, -13395, -13391, -13387, -13383, -13367, -13359, -13356, -13343, -13340, -13329, -13326, -13318, -13147, -13138, -13120, -13107, -13096, -13095, -13091, -13076, -13068, -13063, -13060, -12888, -12875, -12871, -12860, -12858, -12852, -12849, -12838, -12831, -12829, -12812, -12802, -12607, -12597, -12594, -12585, -12556, -12359, -12346, -12320, -12300, -12120, -12099, -12089, -12074, -12067, -12058, -12039, -11867, -11861, -11847, -11831, -11798, -11781, -11604, -11589, -11536, -11358, -11340, -11339, -11324, -11303, -11097, -11077, -11067, -11055, -11052, -11045, -11041, -11038, -11024, -11020, -11019, -11018, -11014, -10838, -10832, -10815, -10800, -10790, -10780, -10764, -10587, -10544, -10533, -10519, -10331, -10329, -10328, -10322, -10315, -10309, -10307, -10296, -10281, -10274, -10270, -10262, -10260, -10256, -10254 };

小迷糊 发表于 2008-6-11 02:09

/**//// <summary>7hb f+Kk @8zc9f
/// 包含汉字拼音的字符串数组。
/// </summary>
private static string[] ps = new string[] ...{ "a", "ai", "an", "ang", "ao", "ba", "bai", "ban", "bang", "bao", "bei", "ben", "beng", "bi", "bian", "biao", "bie", "bin", "bing", "bo", "bu", "ca", "cai", "can", "cang", "cao", "ce", "ceng", "cha", "chai", "chan", "chang", "chao", "che", "chen", "cheng", "chi", "chong", "chou", "chu", "chuai", "chuan", "chuang", "chui", "chun", "chuo", "ci", "cong", "cou", "cu", "cuan", "cui", "cun", "cuo", "da", "dai", "dan", "dang", "dao", "de", "deng", "di", "dian", "diao", "die", "ding", "diu", "dong", "dou", "du", "duan", "dui", "dun", "duo", "e", "en", "er", "fa", "fan", "fang", "fei", "fen", "feng", "fo", "fou", "fu", "ga", "gai", "gan", "gang", "gao", "ge", "gei", "gen", "geng", "gong", "gou", "gu", "gua", "guai", "guan", "guang", "gui", "gun", "guo", "ha", "hai", "han", "hang", "hao", "he", "hei", "hen", "heng", "hong", "hou", "hu", "hua", "huai", "huan", "huang", "hui", "hun", "huo", "ji", "jia", "jian", "jiang", "jiao", "jie", "jin", "jing", "jiong", "jiu", "ju", "juan", "jue", "jun", "ka", "kai", "kan", "kang", "kao", "ke", "ken", "keng", "kong", "kou", "ku", "kua", "kuai", "kuan", "kuang", "kui", "kun", "kuo", "la", "lai", "lan", "lang", "lao", "le", "lei", "leng", "li", "lia", "lian", "liang", "liao", "lie", "lin", "ling", "liu", "long", "lou", "lu", "lv", "luan", "lue", "lun", "luo", "ma", "mai", "man", "mang", "mao", "me", "mei", "men", "meng", "mi", "mian", "miao", "mie", "min", "ming", "miu", "mo", "mou", "mu", "na", "nai", "nan", "nang", "nao", "ne", "nei", "nen", "neng", "ni", "nian", "niang", "niao", "nie", "nin", "ning", "niu", "nong", "nu", "nv", "nuan", "nue", "nuo", "o", "ou", "pa", "pai", "pan", "pang", "pao", "pei", "pen", "peng", "pi", "pian", "piao", "pie", "pin", "ping", "po", "pu", "qi", "qia", "qian", "qiang", "qiao", "qie", "qin", "qing", "qiong", "qiu", "qu", "quan", "que", "qun", "ran", "rang", "rao", "re", "ren", "reng", "ri", "rong", "rou", "ru", "ruan", "rui", "run", "ruo", "sa", "sai", "san", "sang", "sao", "se", "sen", "seng", "sha", "shai", "shan", "shang", "shao", "she", "shen", "sheng", "shi", "shou", "shu", "shua", "shuai", "shuan", "shuang", "shui", "shun", "shuo", "si", "song", "sou", "su", "suan", "sui", "sun", "suo", "ta", "tai", "tan", "tang", "tao", "te", "teng", "ti", "tian", "tiao", "tie", "ting", "tong", "tou", "tu", "tuan", "tui", "tun", "tuo", "wa", "wai", "wan", "wang", "wei", "wen", "weng", "wo", "wu", "xi", "xia", "xian", "xiang", "xiao", "xie", "xin", "xing", "xiong", "xiu", "xu", "xuan", "xue", "xun", "ya", "yan", "yang", "yao", "ye", "yi", "yin", "ying", "yo", "yong", "you", "yu", "yuan", "yue", "yun", "za", "zai", "zan", "zang", "zao", "ze", "zei", "zen", "zeng", "zha", "zhai", "zhan", "zhang", "zhao", "zhe", "zhen", "zheng", "zhi", "zhong", "zhou", "zhu", "zhua", "zhuai", "zhuan", "zhuang", "zhui", "zhun", "zhuo", "zi", "zong", "zou", "zu", "zuan", "zui", "zun", "zuo" };
O5K J!F.wnD'YY*F
/**//// <summary>
/// 包含要排除处理的字符的字符串数组。
/// </summary>
private static string[] bd = new string[] ...{ "，", "。", "“", "”", "‘", "’", "￥", "＄", "（", "「", "『", "）", "」", "』", "［", "〖", "【", "］", "〗", "】", "—", "…", "《", "＜", "》", "＞" };

小迷糊 发表于 2008-6-11 02:10

private static Hashtable _Phrase; B] r~.B+Ec

/**//// <summary>&NAhd:D!ZN5L
/// 设置或获取包含列外词组读音的键/值对的组合。
/// </summary>
public static Hashtable Phrase
...{
getr'h9jV;Mo-y!a1r
...{&J(~$x ~o5yxQ+{
if (_Phrase == null) i"{ N B uBU I(z0E
...{
_Phrase = new Hashtable();
2D,h*r(qMV~Kk/Y
_Phrase.Add("重庆", "Chong Qing");
_Phrase.Add("深圳", "Shen Zhen");
_Phrase.Add("什么", "Shen Me");-s?BtWv h
}
Ahe'C)Gy
return _Phrase;
}Y'h)T#[KaiEux5d

set ...{ _Phrase = value; }D/diE \*xbx;h7X0S
})L+^u6w!W.h u`
Vs/dY*y.y\9^(J
/**//// <summary>
/// 将指定中文字符串转换为拼音形式。n8m f6aj!p;^
/// </summary>6qRo7D,J*[+P
/// <param name="chs">要转换的中文字符串。</param>.a.\wV#N_9[-i~
/// <param name="separator">连接拼音之间的分隔符。</param>
/// <param name="initialCap">指定是否将首字母大写。</param>
/// <returns>包含中文字符串的拼音的字符串。</returns>
public static string Convert(string chs, string separator, bool initialCap)
...{
if (chs == null || chs.Length == 0) return "";2St[2B-F
if (separator == null || separator.Length == 0) separator = "";

// 例外词组
foreach (DictionaryEntry de in CHS2PinYin.Phrase)m/Y-Ot7|
...{
chs = chs.Replace(de.Key.ToString(), String.Format(" {0} ", de.Value.ToString().Replace(" ", separator)));-Mof:qA)L-YU
}3\ X7a5G#L3M*n*Q1R
2\4c-z'Hy!G
byte[] array = new byte[2];
string returnstr = "";B[ V:o5xRA3A
int chrasc = 0;V1g(rO*G_qv-~(sd
int i1 = 0;G:G5b!Ihk+inh
int i2 = 0;
bool b = false;7~R%RQp
char[] nowchar = chs.ToCharArray();0A:R{6|7qw(A1}OE

CultureInfo ci = Thread.CurrentThread.CurrentCulture;{G/ELY^1I ~P
TextInfo ti = ci.TextInfo;9l+uN7f v%E.tc*J

for (int j = 0; j < nowchar.Length; j++)
...{
array = Encoding.Default.GetBytes(nowchar[j].ToString());w)fx1YP~L.j'N
string s = nowchar[j].ToString(); ;ywI%XNyI
2~B`+~}&w.N
if (array.Length == 1)
...{
b = true;,P]"@+^7`0Wx*_
returnstr += s;i4S{JBE!@)ufb
}
else
...{
if (s == "？")
...{
if (returnstr == "" || b == true) returnstr += s;/]9P2tG {b?4d
else returnstr += separator + s;

小迷糊 发表于 2008-6-11 02:10

continue;
} N8D&{VBq7l}W

i1 = (short)(array[0]);+\/lA4zSY1ZW@%D
i2 = (short)(array[1]);V2g/g/V$i+H ij
j&Cj&_O&\
chrasc = i1 * 256 + i2 - 65536;G e:K7My/x#n[1df

for (int i = (pv.Length - 1); i >= 0; i--)/m;^S.N?s7u,`p{)V0R
...{
if (pv <= chrasc)?(^ Z dQ'\o
...{C$b|qg Q
s = ps;
+qK|ux;D,@fDueJ~
if (initialCap == true) s = ti.ToTitleCase(s);!j'?h3B TtP

if (returnstr == "" || b == true) returnstr += s;i|V!T4YP
else returnstr += separator + s;

break; p#p1hC{V I,V
}.o6Mkw:w2a
}chh-F/v0G^
]'[q k+rdze1XP
b = false;
}
}/O c*~-PZ{n

returnstr = returnstr.Replace(" ", separator);
return returnstr;
},]W.Zo%s;F0?;n
z^p$t+vg
/**//// <summary>7M0E1ty(J8b
/// 将指定中文字符串转换为拼音形式。2n_kt-Q%g
/// </summary>
/// <param name="chs">要转换的中文字符串。</param>Q6k{1?6EwE6Wc3Aa
/// <param name="separator">连接拼音之间的分隔符。</param>0s1U#sf:d
/// <returns>包含中文字符串的拼音的字符串。</returns>/^J-MM"V3w.|"r
public static string Convert(string chs, string separator)S2K5ZWb j Td'X
...{
return CHS2PinYin.Convert(chs, separator, false);
}
9O mfMsl/J,aG
/**//// <summary>y(B Ga*C/xB?
/// 将指定中文字符串转换为拼音形式。+N/C~~5N9K^&P3@'n4z7[
/// </summary>]q So8P'om U/F:bI+I
/// <param name="chs">要转换的中文字符串。</param>
/// <param name="initialCap">指定是否将首字母大写。</param>
/// <returns>包含中文字符串的拼音的字符串。</returns>g:z!yk3y|\d+K
public static string Convert(string chs, bool initialCap)
...{
return CHS2PinYin.Convert(chs, "", initialCap);
}I)V9k$b}

/**//// <summary>2}"LSGjv$nn
/// 将指定中文字符串转换为拼音形式。
/// </summary>
/// <param name="chs">要转换的中文字符串。</param>'f;T8z'O,Lg/Z)\:v
/// <returns>包含中文字符串的拼音的字符串。</returns>
public static string Convert(string chs)
...{
return CHS2PinYin.Convert(chs, "");em$D hT2o
}W$ydze2j7s)Z
}

页: [1]

我爱电脑技术论坛's Archiver

C#汉字转拼音功能类